STEM Opportunity Structure 1 - HERI...STEM Opportunity Structure 4 their goals (Link, 2003), provide academic enrichment, offer a sense of guidance and support (Barlow & Villarejo,

STEM Opportunity Structure 1

Running head: STEM OPPORTUNITY STRUCTURE Supporting Future Scientists: Predicting Minority Student Participation in the STEM Opportunity

Structure in Higher Education

Tanya Figueroa, Bryce Hughes, and Sylvia Hurtado University of California, Los Angeles

The National Association for Research in Science Teaching (NARST) April 2013

Rio Grande, Puerto Rico

This study was made possible by the support of the National Institute of General Medical Sciences, NIH Grant Numbers 1 R01 GMO71968-01 and R01 GMO71968-05 as well as the

National Science Foundation, NSF Grant Number 0757076. This independent research and the views expressed here do not indicate endorsement by the sponsors.

Contact: Tanya Figueroa, 405 Hilgard Ave., 3005 Moore Hall, University of California, Los Angeles, CA 90095-1521; Email: [email protected]


Supporting Future Scientists: Predicting Minority Student Participation in the STEM

Opportunity Structure in Higher Education

The United States’ ability to achieve its national goals has historically depended on its

global leadership in science and engineering. The nation is at risk, however, of losing its

competitive edge, demonstrated by recent reports revealing that the domestic investment by

other countries in STEM programs now surpasses that of the United States (National Academy

of Sciences, 2011). As a result recent national imperatives advocate the need to confer an

additional one million STEM degrees over the next decade to ensure the technological

innovation necessary for national competitiveness in a global market (PCAST, 2012). If

postsecondary institutions are to meet national calls to diversify and bolster the STEM

workforce, they must increase the enrollment, retention and graduation of Black, Latino, and

American Indian students. Since students from these groups are among the least likely to

complete a STEM major (HERI, 2010), the National Academy of Sciences (2011) designates

them as underrepresented racial/ethnic minority (URM) students. Improving the persistence of

URM students is also a social justice concern, as unequal hierarchies and oppressive structures

(Lewis, 2001) impede the academic progress and success of URM students in higher education

(Hurtado & Carter, 1997), especially for those who pursue STEM majors (Massey, Charles,

Lundy, & Fischer, 2002).

The first two years of college are critical to college persistence and represent a

particularly vulnerable time for students (Tinto, 1993). Indeed 40-60% of STEM aspirants switch

out of a STEM major within the first two years of taking their first science or math class

(Seymour & Hewitt, 1997). There are a number of reasons why students turn away from STEM.

For some, unsatisfactory grades - attributed to poor academic preparation in math and science

during high school (Henderson & Broadbridge, 2007; Schneider, 2000) and a lack of academic

confidence (Armstrong et al., 2011) – prompt students to withdraw from a particular class or


from school completely (Arendale, 1998). For many others, including those with strong

precollege academic preparatory experiences, poor achievement can be attributed to the

pedagogical practices of introductory classes that weed out students instead of harvesting and

nurturing talent (Seymour & Hewitt, 1997). Notably even successful students feel discouraged

from pursing STEM majors after early encounters with professors who seem unapproachable

and inaccessible or appear to lack an overall ethic of care (Eagan, Figueroa, Hurtado, &

Gasiewski, 2012). These studies suggest that students’ decisions to remain in STEM are in a

large way impacted by the educational environment in which they learn.

URM students in particular face a number of challenges that put them at a disadvantage

when pursuing a STEM major compared to their majority peers that have much to do with

institutional structures that reinforce inequality. According to Massey et al., (2002), URM

students are less likely to have the skills needed to navigate the college environment. They also

are less likely to have attended a quality school and to have had exposure to rigorous academic

preparation in high school that would facilitate the transition into college coursework (Adelman,

2006; Elliott, Strenta, Adair, Matier, & Scott, 1996; May & Chubin, 2003). URM students are

more likely to question their academic ability due to negative stereotypes about URM students

(Aronson & Inzlicht, 2004). Further, the racial climate can make it difficult for URM students to

feel a sense of belonging in college, especially if they are severely underrepresented (Hurtado,

Han, Saenz, Espinosa, Cabrera, & Cerna, 2007; Johnson, 2012), which makes them more

vulnerable to attrition in the face of personal burdens (Inzlicht & Good, 2006; Walton & Cohen,

2007).

Knowing that students contend with a number of barriers to persistence, it is important to

know more about the factors that mitigate the effect of these barriers and help students achieve

and persist in STEM majors. The literature on STEM education demonstrates that

extracurricular and co-curricular activities boost students’ self-confidence, help students define


their goals (Link, 2003), provide academic enrichment, offer a sense of guidance and support

(Barlow & Villarejo, 2004), and grant students the opportunity to become academically and

socially integrated in college life (Astin, 1993). These activities are also associated with greater

academic achievement, retention, and persistence. Such co-curricular activities and programs

comprise what we refer to as the “STEM opportunity structure,” because they support and

enhance students’ participation in a STEM major, serve as pathways into STEM-related

careers, or motivate students to pursue graduate work. Scant research, however, has identified

the individual and institutional factors that affect the likelihood that students will have access to

and/or become involved in these key co-curricular experiences during their undergraduate

career.

The purpose of this paper is to examine predictors of the likelihood that science-oriented

students would participate in or have access to different components of the STEM opportunity

structure: namely undergraduate research programs, supplemental instruction, major-related

clubs or organizations, internship programs, and faculty mentorship and support. This study will

be of interest to STEM faculty and student affairs professionals who play an active role in

providing these STEM-related opportunities to students with the goal of improved academic

success and retention. By identifying the factors that predict access to key college experiences,

practitioners can ascertain whether underrepresented racial minority students and other

students vulnerable to underachievement in STEM are benefiting from these opportunities.

Discrepancies in participation may indicate a need for proactive and early outreach to these

students. This study will also advance the existing body of knowledge about how to facilitate the

success of STEM students, especially those who come from underrepresented backgrounds.

Literature Review

Co-curricular experiences contribute to students’ success in STEM degree programs

and open pathways to a career in STEM or graduate work in STEM. The next section of this


paper will provide a brief overview of the research regarding undergraduate research,

supplemental instruction, major-related clubs and organizations, internships, and faculty

mentoring. Institutions provide these activities to help with STEM retention. This study is

interested in these interventions because they provide both academic enrichment along with

peer and/or faculty support and are therefore part of the STEM opportunity structure.

Undergraduate Research

Of all of the activities within the STEM opportunity structure, the most widely

documented in the higher education literature is undergraduate research. An essential

component of any research program includes the socialization of students into the culture of a

research career thereby increasing students’ confidence and skills as a researcher and

improving their ability to apply their understanding of science in an experiential setting (Kardash,

2006; Kinkead, 2003; Lopatto, 2003; 2004; Russell, Hancock, & McCullough, 2007; Seymour et

al., 2004). Undergraduate research contributes especially to the success of underrepresented

minority STEM students; specifically participation increases their likelihood of degree

completion, of graduating with a high GPA, and enhances aspirations to pursue a STEM career

or graduate degree (Eagan, Hurtado, Chang, Garcia, Herrera, & Garibay, in press; Jones,

Barlow, & Villarejo, 2010; Pender, Marcotte, Sto. Domingo, & Maton, 2010; Strayhorn, 2010).

A study by Hurtado, Eagan, Cabrera, Lin, Park, and Lopez (2008) is presently the only

study that has identified the factors that predict first-year participation in undergraduate

research. Findings from the study indicated that students who participated in first-year

experience courses and pre-professional departmental clubs were significantly more likely to

participate in health science research in their first year of college. Black students were also

more likely to participate in research when they attended institutions that offered undergraduate

research experiences to first-year students as part of a structured program.

Supplemental Instruction (SI)


Supplemental instruction is structured time outside of class that complements a

traditionally difficult course and provides an enriched learning environment in which an instructor

facilitates deeper understanding of course content (Wilson et al., 2011). In contrast to remedial

coursework, SIs target courses covering challenging topics, and participation is voluntary,

avoiding the stigmatization of mandatory remediation (Arendale, 1997). Some research

indicates that students who participate in SI are not significantly different from non-participants

with respect to high school grades (Malm et al., 2010), gender, age, standardized test scores, or

ethnicity (Arendale, 1998); alternatively other research contends that SI participants have

weaker academic backgrounds (Rath et al., 2007) and are more likely to be female (Coletti et

al., 2012) or come from URM backgrounds (Peterfreund et al., 2008).

The benefits associated with SI participation are particularly important for students in

STEM majors (Armstrong et al., 2011; Blat & Nunnally, 2004; Hands, Reid & Younger, 1997)

and provide a positive first experience to students new to the discipline (Malm et al., 2010).

Students who have less confidence in their academic preparation and less exposure to the

material taught in class are the most likely to benefit from the extra academic help provided by

SIs targeting STEM courses (Coletti et al., 2012), with gains being particularly great for URM

students (Peterfreund et al., 2008; Rath et al., 2007). SIs offer a significant tool for STEM

students to overcome challenges resulting from poor academic preparation in high school

(Barlow & Villarejo, 2004). When compared to non-participants, SI participants earn higher final

grades (Armstrong et al., 2011; Blat & Nunnally, 2004; Rath et al., 2007) and are less likely to

repeat a course (Peterfreund et al., 2008). Since passing introductory STEM classes is

necessary for degree progress and the completion of a STEM degree, it is no surprise that SI

participants are more likely to progress to subsequent STEM classes in a sequence,

(Peterfreund et al., 2008), complete their first year of college (Malm et al., 2010), and graduate

in science majors (Rath et al., 2007).


Major-Related Clubs

Participation in STEM-related student clubs or organizations enhances students’

educational experiences and is important to the career development of students (Bohlscheid &

Clark, 2012). Durham and Marshall’s (2012) study on engineering college students

demonstrated that major-related clubs provided students with opportunities to befriend,

socialize, and study with other members and thereby access more social and academic support,

ask questions, and network with professions in their field. Students participating in the

engineering clubs stated that they more enjoyed learning, had higher motivation and self-

confidence, and more clear educational and career goals as a result of participating in these

clubs (Durham & Marshall, 2012). Involvement in major related clubs also gives students

access to networks of people who can teach them the rules of science participation (Do et al.,

2006).

Another benefit of participation in clubs and internships related to one’s STEM major, is

that it provides students with critical information on issues facing the profession (Bohlscheid &

Clark, 2012). Further, students holding leadership positions in STEM-related clubs develop

interpersonal skills and gain skills in time and conflict management, all of which are transferable

in the job market and better prepare students for entry-level positions after college (Durham &

Marshall, 2012). This might explain why students who participate in clubs and internships tend

to secure jobs sooner after graduation (Sagen et al., 2000). Joining a major-related club or

organization also has a strong influence on the likelihood that URM students will continue to

have an interest in a STEM career (Herrera, Hurtado, & Chang, 2011). Several studies show the

importance of participation in departmental clubs for the retention for women of color (Espinosa,

2011) and underrepresented groups (Chang, Hurtado, Sharkness, and Newman, in review).

Internships


Participation in hands-on work experiences related to one’s respective STEM discipline

enhances the academic experience (Perna, Cooper, & Li, 2007) as students apply what they

have learned in the classroom to real-world practical situations and gain an improved

understanding of their particular field (Jaeger et al., 2008). Involvement in internships may also

give students an opportunity to feel like they are contributing meaningfully to the field, help

students learn the norms of science, and decrease feelings of marginalization by integrating the

student into the STEM community (Hunter, Laursen, & Seymour, 2007). In one study on

engineering students, internships were found to support an engineering student identity,

enhance students’ self-confidence, and increase their motivation to finish school (Do et al.,

2006).

STEM-related work experiences, like cooperative education, can also strengthen a

student’s commitment to their major and have a positive influence on STEM persistence if the

student decides that their major coincides with their career interest (Jaeger et al., 2008).

Students who participated in cooperative education (i.e. a short-term real-world professional

work experience removed from the university environment) had higher final GPAs and were

more than five times more likely than non-participants to persist in their STEM major (Jaeger et

al., 2008). Internships can additionally be a gateway to gainful employment after graduation for

STEM students (Do et al., 2006), and URM students in particular (Inroads, 1993), as they help

students build a professional network (Do et al., 2006), and better prepare students for their first

job in STEM (Jaeger et al., 2008).

Faculty Mentoring and Support

Although a plethora of studies examine student-faculty interactions broadly, we do not

consider happenstance and haphazard interactions to be part of the opportunity structure

because these types of interactions do not have a significant impact on students’ academic and

career trajectories in STEM. Instead, this study is interested in deliberate mentoring and support


from faculty, in which faculty sponsor (i.e. provide letters of recommendation), coach, role

model, and counsel (Kram, 1983) students. Indeed researchers have linked faculty mentorship

and support to increased student GPAs (Strayhorn, 2009), improved performance in

introductory STEM courses (Tsang et al., 2006), and improved overall retention in science

(Packard, 2004). The mentoring relationships students establish with their undergraduate

research advisors also have a great impact on the effectiveness of undergraduate research

experiences (Thiry & Laursen, 2011).

Although there are a variety of studies exploring the benefits of faculty mentorship, only

one examines the factors that predict the likelihood of students receiving such support (Eagan,

Herrera, Garibay, Hurtado, & Chang, 2011). Findings demonstrated that high achieving students

and students who were already more likely to be successful in STEM (such as those who enter

college with high SAT scores, have graduate school aspirations, or a strong science identity)

were most likely to have access to faculty mentoring and support. As a result of the findings, the

authors encourage faculty to reach out and provide mentorship to a wider pool of students,

especially those that don’t fit the typical notion of a “high-achieving” student.

Conceptual Framework

We draw from frameworks regarding goal setting, social capital, and institutional context

to investigate the factors that promote and reduce a student’s likelihood of engaging in different

experiences within the opportunity structure during college.

Individual Factors

Various student demographic characteristics impact the degree to which students

engage in certain activities in college. Students who demonstrate high levels of academic

achievement, aspirations, and savvy have a greater propensity to seek out and access

opportunities that will help them achieve their career or educational goals (Green & Bauer,


1995; Wanber, Welsh, & Hezlett, 2003). This phenomenon is known as the “rising star”

hypothesis (Ragins, 1999; Singh, Ragins, & Tharenou, 2009).

Individual effort alone however does not explain why some students are highly involved

in college and why others are not (Braxton, 2000; Tierney, 1992). URM undergraduate students

in particular confront multiple structural barriers to academic success including little explicit

guidance on how to navigate the college environment (Massey et al., 2002), negative

perceptions of underrepresented racial groups (Schwitzer, Griffin, Ancis, & Thomas, 1999), and

inadequate financial aid prompting serious concerns about financing college (Hurtado, Han,

Saenz, Espinosa, Cabrera & Cerna, 2007). Many of these barriers are also reflected within the

social and institutional factors that either ease or restrict access to the structure of opportunity

for different students.

Social Factors

Students with expansive social networks are more likely to identify and access activities

within the opportunity structure. Indeed institutional agents (like faculty and administrators)

equip students with the knowledge, social networks, resources, and experiences to which they

would not typically have access (Stanton-Salazar, 1997; 2001; 2010). Stanton-Salazar’s (2010)

Institutional Agents framework acknowledges that there are inequalities in access to valued

resources and opportunities due to differences in social capital (Stanton-Salazar, 1997) and

cultural capital, which are largely influenced by class, race, and gender (Bourdieu, 1986,

Bourdieu & Passeron, 1977). This theory illuminates the privileged position of some students in

relation to others and demonstrates the importance of these support agents in empowering and

advocating for students, and broadening access to opportunities for students from

disadvantaged social positions. Faculty, as the primary agents of socialization, also have the

ability to orient students to institutional and disciplinary norms and the acceptable dispositions


within science (Becher, 1989), thereby better positioning students to navigate higher education

institutions.

Institutional/Structural Factors

Various institutional characteristics including selectivity, size, research orientation, and

type shapes the availability of structured opportunities available to students (Porter, 2006), the

extent to which institutions are culturally responsive to student needs (Outcalt & Skewes-Cox,

2002), and the culture around student engagement (Merkel, 2003). Ultimately institutional

contexts matter, and matter greatly, with regard to how productive institutions are relative to

others in engaging students, integrating them into campus life, and developing their talent in

STEM (Porter, 2006; Chapman & Pascarella, 1983; Hamrick, Schuh, & Shelley, 2004).

Methods

Data and Sample

The data for this sample is drawn from the Cooperative Institutional Research Program

(CIRP) 2004 Freshman Survey and follow-up 2008 College Senior Survey. The Freshman

Survey (TFS) is administered to first-time freshmen students during freshman orientation or

during their first term in college and collects demographic information and information about

students’ precollege experiences, attitudes, values, goals, self-perceptions, and expectations for

college. College seniors complete the College Senior Survey (CSS) in the spring of their fourth

year, and this instrument collects information about the experiences students had while in

college as well as their self-perceptions, values, attitudes, career aspirations, and post-

graduation plans. The longitudinal response rate for the 2004 TFS and 2008 CSS was

approximately 23%. Response weights were calculated to adjust for potential non-response

bias. The full longitudinal dataset includes information from 6,224 students at 238 institutions.

Institutional data for the 2006-2007 academic school year was taken from the Integrated

Postsecondary Education Data System (IPEDS) and merged into the longitudinal data set.


Eagan et al. (in press) provides more information on the sampling and weighting strategies

applied to this dataset. The final analytic sample included 4,046 students attending 212

institutions who indicated that they aspired to a STEM major at the start of their undergraduate

studies.

Variables

This study analyzes five dependent variables: 1) participated in an internship program,

2) participated in an undergraduate research program, 3) joined a club or organization related to

one’s major, 4) had instruction that supplemented coursework; and 5) frequency study received

faculty mentorship and support. For dependent variables one through three, students had the

option of marking yes or no to indicate whether they participated in the respective activity. For

the fourth dependent variable, students marked how often they had instruction that

supplemented coursework (not at all, occasionally, or frequently). The faculty mentorship and

support construct measured the extent to which students and faculty interacted in relationships

that fostered mentorship, support, and guidance, in both academic and personal domains. This

score was determined by CIRP using item response theory, and items included students’

reports of the frequency with which faculty provided nine support activities (see Appendix B for a

list of these items). Responses were on a three-point scale: not at all, occasionally, and

frequently. A higher score on faculty mentorship indicated a greater degree of a mentoring

relationship with faculty. (See Sharkness, DeAngelo, and Pryor, 2010, or CIRP Construct

Technical Report, 2010, for more information on the creation of the various constructs used in

this paper.)

To analyze each of the five outcomes described above, we relied on a common set of

predictor variables. Prior literature on student engagement and our conceptual frameworks on

goal setting, social capital, and institutional context guided selection of the variables used in the

models. In our analyses, we added variables in conceptually related, temporally sequenced


blocks. First, we included student demographic characteristics (e.g., sex, racial/ethnic

background, and socioeconomic status) in the models. Next, we added several pre-college

measures (e.g., prior academic preparation, high school activities, and degree aspirations) to

the models to see if any observed differences between groups could be accounted for by

differences in these areas. Third, we added students’ college experiences and attitudes

measured on the CSS to the models. Finally, institution-level variables were added in the last

block, and these measures included structural characteristics of the institution such as size,

selectivity, and control. We ran identical models for each of the five dependent variables, with

the exception that we used each dependent variable as an independent predictor in the models

for the other four outcomes. Appendix A contains a complete list of variables in the analysis and

their corresponding coding schemes, and Appendix B provides the individual items included in

the model.

Analysis

Missing Data

In order to maximize the sample available for analysis, missing data were replaced,

wherever appropriate, in a multi-step process. First, we removed from our sample all students

who had missing data on one of the dependent variables and students who were missing

information on key demographic characteristics such as gender, race, or native language. In

total, 19 students were missing information in one or more of these areas (< .5%). For the

remaining variables of interest, we analyzed the extent to which missing data occurred. Overall,

there was very little missing data; only three variables had more than 3% of its cases missing.

The variable with the highest proportion of missing data was students’ degree aspirations at

college entry at 9.4%, followed by parental income at 8% (used to create the SES factor) and

SAT composite score at 6.5%.


Given the relatively few instances of missing data across the variables used in the

analysis, we imputed missing data using the expectation-maximization (EM) algorithm in SPSS.

The EM algorithm employs maximum likelihood estimation techniques to impute values for

cases with missing data. Because EM uses most of the information available in the dataset to

produce the imputed values, it is a more robust method of dealing with missing data than

listwise deletion or mean replacement (Allison, 2002; Dempster, Laird, & Rubin, 1997;

McLachlan & Krishnan, 1997). Distributions of variables were compared before and after

missing values were imputed, and were found to be virtually identical.

Multi-level Analysis

The clustered, multi-level nature of our data necessitated the use of hierarchical linear

modeling (HLM) for the continuous dependent variables and hierarchical generalized linear

modeling (HGLM) for the dichotomous dependent variables. Performing single-level analyses

with multi-level data can underestimate the standard errors of model parameters, increasing the

likelihood of committing Type-I statistical errors (de Leeuw & Meijer, 2008; Raudenbush & Bryk,

2002). To ensure the use of multi-level techniques was justified, a fully unconditional model (i.e.,

a model with no predictors) was run to assess the extent to which our outcomes varied across

the institutions in our sample. For the dichotomous outcome variables (participated in research,

a major-related club, and internships), the null models showed that the between-institution

variance components for all three outcomes significantly varied across institutions. Given this

significant variation and our interest in the examining how institutional contexts both directly

affect students’ likelihood of participation in the opportunity structure and moderate individual-

level relationships, we proceeded with the use of HGLM. For the continuous outcome variables

(i.e. faculty mentorship and participation in SI) we calculated the amount of the variance that

was due to differences between institutions. For faculty mentorship, roughly 11.7% of the

variation in scores is attributable to differences between institutions. Although only 2.4% of the


variation in students’ frequency of participation in SI was attributable to institutional differences,

we proceeded with HLM analyses for this outcome due to our interest in examining the role of

institutional context.

Since we were interested in the average effect of each predictor on students’

participation across the five activities in the opportunity structure that we examine, we chose to

grand-mean center all continuous and ordinal variables. Dichotomous variables were left un-

centered because a zero value on these variables is meaningful.

Limitations

One of the primary limitations with this data set is that it only includes the responses of

students who persisted to the fourth year of college. The students in this dataset, who

successfully managed to stay in college until their senior year, were probably very different from

the students who aspired to major in STEM during their freshman year of college but transferred

to another university or dropped out of school completely. Regrettably, no data is available on

these students. There is a possibility that student dropouts or transfers had differential patterns

of participation in the various activities in the opportunity structure and had different college

experiences than those who persisted at the same institution after four years.

A related limitation of our study is that the CSS had a relatively low longitudinal response

rate (23%), and thus the extent to which our results can be generalized to a larger group of

students may be limited. Although we attempted to correct for non-response bias that may have

been introduced by the low response rate, our correction was necessarily limited to the

information we had available, and we may not have taken all of the important factors into

consideration. Also, a number of the independent variables in this study are self-reported (i.e.

GPA, SAT scores) and it is possible that students’ answers do not accurately reflect what

actually occurred. Previous research however demonstrates high overall validity of self-

reported scores on academic performance (Cole & Gonyea, 2010).


Finally, as our dependent variables were taken from the 2008 CSS, our dependent

variables were measured at the same point in time as many of our independent variables.

Therefore we cannot assume a causal relationship between the dependent variables and those

independent variables measured in 2008. Our purpose is to identify the experiences that are

associated with a greater or lesser probability of participating in the STEM opportunity structure;

thus the establishment of causation is not necessary to address the focus of our study.

Results

Descriptive Statistics

Table 1 provides descriptive statistics for variables in the model. Roughly 40% of

students self-identified as White, 16% as Latino, 15% as Black, 12% as Asian, and 15% as

belonging to two or more racial groups. Less than 1% of students identified as American Indian

or indicated “other.” Approximately 62% of the students were female. The results show that

46% of students who aspired to major in STEM at college entry participated in an internship

program by the fourth year of college, 20% participated in an undergraduate research program,

and 60% joined a club or organization related to their major. The average faculty mentorship

score for STEM aspirants was 48.43 (which is slightly below the population average for faculty

mentorship) and 85.6% of students had instruction that supplemented their coursework at least

occasionally.

HLM and HGLM Models

We begin by presenting the HLM results for supplemental instruction and faculty

mentorship, found in Tables 2 and 3, respectively. We summarize the results for the three

dichotomous dependent variables in Table 4. For significant parameter estimates in Table 4 we

have corresponding delta-p statistics, which represent the expected change in probability

associated with a one-unit increase from the mean of the independent variable (Cruce, 2009;

Petersent, 1985).


Supplemental Instruction (SI)

Demographic characteristics account for 0.68% of the student-level variance in the

frequency of participation in SI. This proportion rises to 1.8% after adding students’ precollege

experiences into the model and jumps to 16.1% once college experiences are accounted for.

The level-one variance drops to 5.74%, however, once institutional-level variables are added to

the model, which indicates a poorer fit model at the student-level. Further, none of the

institutional variables reaches significance in the final model, meaning differences in the

probabilities of participating in supplemental instruction is not explained by attendance at

different types of institutions. Although we do not discuss the results from the model containing

institutional variables, we include the results from this model in Table 2 so that readers can see

the coefficients and p-values for the various institutional variables that we examined. Below is a

discussion of the best-fit model, which accounts for 16.1% of the student-level variance.

Students who are non-native English speakers utilize supplemental instruction more

frequently than native English speakers; no other demographic characteristics are significant.

It’s possible that some of the assistance non-English native speakers receive in SI is related to

learning new concepts in a language they are still learning. Further, students who had

participated in summer research programs during high school tend to participate in SI more

often than those who do not. A handful of college experiences also seem to matter in predicting

how frequently students utilize SI. Students who work on independent study projects, study with

or tutor their peers, or ask a professor for advice outside of class tend to engage more

frequently in supplemental instruction. Further, joining a club or organization related to one’s

major, participating in an academic program for racial/ethnic minorities, and spending more

hours per week socializing with friends are associated with more frequent use of supplemental

instruction. Perhaps peers influence each other to seek out the resources needed to succeed

academically. Student perceptions of faculty also matter: higher agreement with the statement


that faculty at the institution care about students’ academic problems is associated with more

frequent use of SI. Students who are more academically disengaged also more frequently use

SI. This is a curious finding given that we would expect a student who enters coursework more

disengaged to be less likely to seek out supplemental instruction. This finding indicates that

students who are disengaged and unsuccessful in the regular classroom have to seek additional

help in supplemental instruction. Thus SI serves both students who are really engaged and

students who are not very involved in the regular science classroom. Finally students who are

more satisfied with the courses in their major, have a higher sense of belonging, have more

positive cross-racial mentorships, and receive higher levels of faculty mentorship tend to more

frequently engage in supplemental instruction.

Faculty Mentoring and Support

Demographic variables explain a very small percentage (0.57%) of the student-level

variation in the frequency with which students receive faculty mentorship, which increases

slightly to 3.72% with the addition of precollege experiences into the model. Including college

experience variables to the model increases the proportion of level-1 variance explained by the

model to 50%, which then drops to 43.9% when institutional variables are added. A number of

institutional variables significantly predict the frequency with which student received faculty

mentoring. Students attending smaller institutions, master’s comprehensive universities

(compared to liberal arts colleges), and HBCUs tend to receive more frequent faculty

mentorship. Students also tend to receive more frequent faculty mentorship at institutions where

the student body agree more strongly that the faculty at their institution are interested in

students’ personal problems. Students at institutions that have a higher proportion of White

students and lower proportion of STEM students also have higher faculty mentorship scores;

this was over and above institutional size and HBCU status.


Although there are no significant differences by gender in the final model, we did detect

significant differences by race/ethnicity with regard to the frequency with which students receive

faculty mentorship and support. Multiracial students appear to have less frequent mentorship

activity from faculty than their White peers. Further, a greater concern about financing college

and higher SAT scores predict receiving less mentorship from faculty. Students who spent more

hours per week talking to teachers outside of class in high school also have more frequent

mentorship from faculty in college, likely because they continue this behavior in college, a

finding that supports the rising star hypothesis. Student aspirations at college entry also have a

significant impact on the extent to which they receive mentorship from faculty. Compared to

their peers who aspire to a bachelor’s degree, students who are interested in an associate’s

degree or no degree have less frequent mentorship from faculty, while those who aspire to a

master’s degree have more frequent mentorship interactions with faculty. The latter finding is

important because students who plan to apply to graduate school need guidance and

sponsorship from faculty to increase their chances of gaining admission into top programs.

The college environment and a number of college experiences also matter when it

comes to the extent to which students receive mentoring from faculty. Students majoring in

engineering and mathematics have less frequent mentoring and support from faculty than their

peers who switch to non-STEM majors—a finding with important implications for retention in

these majors. Having a higher college GPA and placing a higher importance on the discovery

or enhancement of knowledge to one’s career path are both associated with more frequent

faculty mentorship. Students who at some point in their college career enroll in an honors or

advanced college course, participate in a program to prepare for graduate school, or present

research at a conference typically receive more frequent faculty mentorship. Students who

more often work on independent study projects, meet with an advisor or counselor about their

career plans, ask a professor for advice outside of class, or utilize supplemental instruction, tend


to be the recipients of more frequent faculty mentorship and support. Since many of these

activities are help-seeking behaviors, it is not surprising that they relate to receiving more

frequent support from faculty. Alternatively students tend to receive less faculty mentorship if

they more frequently feel intimidated by their professors or feel isolated from campus life.

Students’ opinion of the faculty in their institution also significantly associates with the extent to

which they receive mentoring from faculty: students who more strongly agree with the

statements that faculty at their institution are interested in both students’ academic and personal

problems tend to receive greater mentorship. This finding was expected as students who

receive more support ought to perceive greater faculty concern for their academic welfare.

Students who report greater satisfaction with the racial/ethnic diversity of the student body and

with the courses in their major field tend to enjoy more frequent mentoring from faculty. Finally

students who are more academically disengaged tend to receive less faculty mentoring.

Although we can’t be sure, it is possible that students start to disengage in college after having

experienced a lack of faculty attention.

Internship Programs

Two institutional variables significantly predict students’ likelihood of participating in an

internship program. Specifically, a 100-point increase in the average SAT score of an

institution’s incoming freshman class (indicating greater institutional selectivity) increases STEM

aspirants’ likelihood of participating in an internship program by 4.51%. Even more pronounced

is the effect of institutional control; enrolling in a private as opposed to a public institution

increases the probability of participating in an internship program by 13.89%.

In addition to institutional characteristics, several background attributes significantly

predict STEM aspirants’ probability of participating in an internship. Students from higher SES

backgrounds are more likely to participate in an internship program. Students are less likely to

participate in an internship program if they place a greater importance on enrolling in college to


prepare for graduate or professional school (-5.21%) or score higher than average on the SAT

(-2.53%). Internship participation may be strongly linked to students’ goals, with students having

aspirations for graduate school and higher SAT scores potentially having career or educational

goals that do not necessitate participation in internships. There are no differences by race or

gender in the likelihood of participating in an internship program.

We saw significant variation in participation in internships across student major.

Students in majors not included in our broader aggregated categories but still classified as

STEM (7.71%), and engineering students (26.57%) are more likely to participate in an internship

than their peers who switch to a non-STEM major. This suggests that internships are more

characteristic of the training students receive in engineering – perhaps because internships are

connected to the acquisition of jobs - compared to the training received in other major field.

Alternatively those who major in a health profession program (i.e. nursing, pre-med, pre-

pharmacy, pre-dental, pre-vet) are 12.05 percentage points less likely to participate in an

internship program than students who switch to non-STEM majors. Further, higher average

grades during college are associated with a greater probability (2.81%) of participating in an

internship program, which may be due to a selection process in which the more competitive

internships take only the highest achieving students.

Student experiences in college have particular salience in predicting whether they

participate in an internship program. Students who fail one or more courses (-10.42%) or work

full-time while attending school (-6.56%) are less likely to participate. Alternatively, students

have higher probabilities of participating in an internship if they enroll in honors or advanced

level college courses (7.18%), participate in an academic program for racial/ethnic minorities

(10.05%), join a club or organization related to their major (10.9%), or present research at a

conference (10.11%). Spending more hours per week working for pay off campus (1.03%) and

career planning (7.25%) is associated with a higher likelihood of participating in an internship


program, which is an expected outcome given that some internships are paid. Interestingly,

students who more strongly agree that faculty at their institution are interested in students’

academic problems (-4.46%) and who are more satisfied with the racial/ethnic diversity of the

student body (-2.87%) are less likely to participate in an internship. Further exploration will be

necessary to explain these preliminary findings. Finally every one-standard deviation (S.D.=10)

increase from the mean in students’ academic self-concept predicts a 4.48 percentage point

increase in students’ probability of participating in an internship.

Undergraduate Research Program

The same institutional variables that predict participation in an internship also predict

participation in an undergraduate research program: a 100-point increase in the average SAT

scores of an institution’s incoming freshman class is associated with a 4.55% increase in

likelihood that a student participates in a research program. In contrast to the likelihood of

having an internship, students at private institutions are 8.65% percent less likely to participate

in research programs in comparison to their peers attending public colleges and universities.

Student demographics and background characteristics also matter. Black students are much

more likely (16.27%) than White students to participate in undergraduate research programs.

This finding is especially encouraging given previous research on the importance of

undergraduate research for the persistence of underrepresented racial minority students in

STEM (Russell et al., 2007). Students whose parents do not work in a STEM-related occupation have

a higher probability (4.6%) of participating in research programs. High school academic

preparation and achievement also play significant roles in students’ chances of being engaged

in research programs. Students’ probability of participating in a research program increases by

2.58% for every 100-point increase in their SAT score from the mean. Oddly, students who took

more years of physical science courses in high school are less likely (-1.54%) to participate in

an undergraduate research program during college. Degree aspirations are a significant and


large predictor of participation in a research program. Specifically, students who upon college

entry reported that they aspired to attain a doctoral degree are 9.51% more likely to participate

in a research program.

Student major also significantly predicts participating in an undergraduate research

program. Students majoring in the biological sciences (13.89%) and the physical sciences

(23.54%) have much higher probabilities of engaging in an undergraduate research program

than their peers who switch to a non-STEM major. Alternatively, students majoring in a health

profession field are 9.93 percentage points less likely to engage in an undergraduate research

program. Difference in participation in research may be in part due to differences by department

in faculty involvement in research programs; departments that are funded by external grants

from NIH and NSF may offer more opportunities for students to participate in research. Further

a greater importance placed on the discovery or enhancement of knowledge to one’s planned

career and a higher college GPA are associated with increased chances of having participated

in an undergraduate research program. A selection process that exists on some campuses, in

which students with higher GPAs are more likely to be selected to participate in undergraduate

research programs, may explain the GPA finding.

Various college experiences also significantly predict the likelihood of participating in a

research program. Working full-time while attending school (-7.98%) is associated with a

reduced likelihood of participating in an undergraduate research program. In contrast, students

who participate in a program to prepare for graduate school (20.61%) and present research at a

conference (35.02%) tend to have a greater likelihood of engaging in an undergraduate

research program. It may, however, be the case that research participation later leads to

presenting research and preparing for graduate school, instead of vice versa. Additionally, the

more frequently students work on an independent study project the more likely (6.12%) they are

to engage in undergraduate research. Student perceptions also contribute to the likelihood that


students engaged in research. Specifically a one-unit increase in agreement (signifying greater

agreement) with the statement that faculty at the institution are interested in students’ academic

problems is associated with a 5.06 percentage point increase in the probability of participating in

a research program. Conversely a one-unit increase in agreement in the statement that there is

strong competition among most students for high grades at the institution is associated with a

2.6% decrease in the odds of engaging in an undergraduate research program. Previous

research has linked perceived competitiveness of the campus academic environment with more

difficulty in adjusting and transitioning to the science environment for first year students

(Hurtado et al., 2007) and diminished STEM persistence for students (Hurtado, Eagan, &

Hughes, 2012); perhaps some of this effect results from having to compete with other students

for limited opportunities like undergraduate research.

Major-related club or organization

Institutional selectivity is the only institutional variable that significantly predicts a

student’s likelihood of joining a major-related club or organization. Specifically a 100-point

increase in the average SAT score of the incoming freshman class is associated with a 7.98%

decrease in students’ probabilities of participating in a major-related club. Stated another way,

students at less selective institutions are more likely to participate in major-related clubs and

organizations than those at more selective institutions. Students who had higher high school

GPAs and research experience prior to college have an advantage in participating in a major-

related club: a one-unit increase in high school GPA is associated with a 3.98% increase in

one’s odds of participation. Students who during high school participated in a health science

research program sponsored by a university are 14.47 percentage points more likely to join a

major-related club.

Although working full-time decreases students’ odds of joining a major-related club by

6.36%, enrolling in honors or advanced college courses (5.66%), presenting research at a


conference (8.33%), participating in an internship program (10.77%), taking part in a program to

prepare for graduate school (7.77%), and being involved in an academic program for

racial/ethnic minorities (8.63%) all increase students’ likelihood of joining a club or organization

related to their major. A one-unit increase in the frequency (indicating a greater frequency) with

which students discuss course content outside of class (6.87%), tutor another college student

(4.69%), ask a professor for advice outside of class (5.32%), or utilize supplemental instruction

(4.07%) augment the odds a student participates in a major-related club. Time spent on various

activities also significantly predicts students’ chances of participating in a major related club.

Students who spend more time commuting to campus are less likely (-2.37%) to participate in

major-related clubs. Alternatively, students who spend more time talking with faculty outside of

class (3.69%) or career planning (2.13%) are more likely to participate in these clubs or

organizations. Students who more strongly agree with the statement that there is strong

competition among students for high grades are more likely (4.64%) to join a major-related club.

Finally, a stronger sense of belonging is associated with a higher probability of joining a major-

related club: specifically a one-standard deviation increase in sense of belonging is associated

with a 3.58% increase in the chances that students are involved in such a club. This could be

evidence that major-related clubs and organizations, like other opportunities for student

involvement with peers and faculty, create a sense of community and increase students’ sense

of belonging on campus.

Discussion and Conclusion

Activities that boost student’s self-confidence, help them define their goals (Link, 2003),

provide academic enrichment, a sense of guidance and support (Barlow & Villarejo, 2004), and

the opportunity to become academically and socially integrated in college life (Astin, 1993) are

associated with greater academic achievement, retention, persistence. Such activities comprise

what we have termed the “STEM opportunity structure,” because they support and enhance


students’ participation in a STEM major, and serve as pathways to STEM-related careers or

motivate students to pursue graduate work. With this in mind there, are several findings from

our study that merit discussion.

First, out of all the variables we tested, the most reoccurring college experience that

was detrimental to students’ ability to participate in the opportunity structure is working full-time

while attending school. Indeed, working full-time while attending school significantly reduces the

chances that students participate in an internship program, engages in an undergraduate

research program, or joins a major-related club or organization. Previous research offers an

explanation as to why this is the case: more time spent in paid employment means less time to

participate in co-curricular opportunities that integrate students into college life (Pascarella &

Terenzini, 2005; Riggert et al., 2006). When considering this finding in light of other findings

from our study—namely that students from lower socioeconomic backgrounds are less likely to

participate in internship programs and more serious concerns about one’s ability to finance

college translate to less frequent faculty mentorship—it becomes clear that financial barriers are

a considerable issue for some students. As outlined in our conceptual framework, we fully

expected financial concerns to be a barrier for underrepresented students (Hurtado, Han,

Saenz, Espinosa, Cabrera & Cerna, 2007); however, the findings from this study indicate that

financial issues burden students more broadly. Since careers in STEM tend to be among the

most prestigious and lucrative, it is a shame that students pursuing STEM as a means to

upward social mobility are the very students that may be the least likely to access activities that

support their success and increase their chances of securing a career in STEM. Thus,

institutions must investigate how they can better support students who likely have the capability

to succeed in college, but have unmet financial need. It might be the case that students who

work full-time are unaware of the financial assistance they are eligible to receive or do not know

how to apply for it, and therefore are not taking advantage of the available aid (Torres et al.,


2010). The shift from need-based aid to merit-based aid is worrisome given that low-income

students also tend to be academically average students (Torres et al., 2010), which has great

implications for socioeconomic equity in access to higher education and the opportunity

structure in STEM on college campuses.

Second, entering college with higher degree aspirations set the stage for future

engagement in the opportunity structure. Specifically, all else being equal, students who aspire

to attain a graduate or professional degree are more likely to participate in the co-curricular

activities we investigated. Contrary to expectations, however, pre-college academic

achievement is not a consistent positive predictor of participation in the opportunity structure.

For example although SAT score is a positive predictor of participation in an undergraduate

research program, having higher SAT scores reduces the likelihood that students will be

involved in an internship program and reduces the frequency with which students receive

mentorship from faculty. Differential involvement in internships, however, may be indicative of

the different types of goals students possess and the types of internships offered on a college

campus.

Although pre-college academic achievement doesn’t appear to make a significant

difference in participation in the opportunity structure, academic achievement once in college

matters a great deal. Specifically, high achieving college students (as measured by overall

college GPA) are notably more likely to participate in an internship program or an

undergraduate research program and more frequently receive faculty mentorship and support.

Similarly, students who enroll in honors or advanced courses are significantly more likely to

participate in an internship program or join a major-related club or organization, and also tend to

receive more frequent faculty mentorship. These findings suggest that students who are already

best prepared and possess the requisite social capital for STEM success are more likely to

access the structure of opportunity in STEM. This is consistent with the rising star hypothesis -


the assertion that the best-prepared and high-achieving students will be the most likely to

engage in these activities, (Singh, Ragins, & Tharenou, 2009). Perhaps institutions and

institutional agents should better leverage their resources to support the talent development of

STEM aspirants, as participation in activities within the opportunity structure appear to be driven

by selection effects that prefer only the most talented students. Further, perceptions of

competition diminish the involvement of students who can do well in science, have great STEM

aspirations and ambition, and could thus benefit greatly from these opportunities.

Third, when all else is held constant, demographic variables in the final models did not

seem to have a consistent and large predictive role in determining whether or not, or the extent

to which students participated in various activities within the educational opportunity structure. It

is noteworthy that although demographics aren’t significant, experiences and traits that tend to

be correlated with certain demographic characteristics are determinative of students’

participation in the opportunity structure.

Fourth, we saw demonstrated differences in participation in the opportunity structure by

major: engineering students have higher probabilities of participating in internships than

students who switch out of STEM, which is expected given that an engineering bachelor’s

degree is a professional degree and thus those pursuing one are likely to be career-oriented

(IEEE, 2007). By contrast, biological and physical science students have a greater likelihood of

engaging in undergraduate research, which corresponds to a greater likelihood of pursuing

graduate school within these fields (Hathaway, Nagda, & Gregerman, 2002). Engineering and

math majors tend to receive a lesser amount of faculty mentorship compared to STEM aspirants

who later switched out of STEM. Students in health professional majors are more likely to join a

major-related club or organization, which may be indicative of the plethora of student chapters

of professional associations in the health sciences on today’s campuses.


Fifth a handful of key college experiences open the door to increased participation in the

opportunity structure. These college experiences include being involved in independent student

projects, partaking in graduate school preparation programs, participating in academic programs

for racial/ethnic minorities, and presenting research at a conference, all of which are associated

with an increased participation in at least three of the five dependent variables. It is important to

keep in mind however, that involvement in these college experiences may have occurred as a

result of participating in other activities within the opportunity structure. That is many of the

STEM interventions are designed to capture these activities. Moreover, faculty and other STEM

practitioners are often involved in activities available in the opportunity structure. Participation in

these activities thus provides students more intimate access to institutional agents who can

assist in their socialization into the STEM discipline (Stanton-Salazar, 1997; 2001; 2010). These

experiences also help students develop a broad support network of peers and thereby help

students cultivate the requisite cultural and social capital needed to succeed in STEM.

Given the poor retention and graduation rates of underrepresented minorities in STEM

disciplines (HERI, 2010), it was encouraging to find that involvement in an academic program

for racial/ethnic minorities promotes student participation in internship programs, membership in

clubs or organizations related to one’s major, and more frequent utilization of supplemental

instruction. Academic programs for racial/ethnic minorities may open the doors to other

beneficial opportunities by helping URM students overcome the challenges that they are more

likely to face such as negative stereotypes (Aronson & Inzlicht, 2004) and little knowledge of

how the college system functions and how to navigate it (Torres, Gross, & Dadashova, 2010).

These programs may also provide crucial social support to students who participate in them.

Although, our variable measuring involvement in academic programs for racial/ethnic minorities

does not provide any specific information about the type of support students receive in these

programs, our findings demonstrate that academic programs targeting racial/ethnic minorities


increase involvement in activities that support STEM success. Future research should

investigate which elements of these academic programs propel participants to become

increasingly engaged. It would also be interesting to determine whether these programs have a

differential impact on students by race, gender, or SES.

Finally, given our interest in the examining how institutional contexts impact participation

in the opportunity structure, it is informative to find that different types of institutions appear to

better facilitate student involvement in activities within the opportunity structure compared to

others. Specifically, private institutions and more selective institutions are better positioned to

promote participation in internship programs. It’s possible that these types of institutions, which

already have vast amounts of institutional resources, have strong relationships with local

industry that ease placement. More selective public institutions alternatively increase the odds

that students participate in undergraduate research programs. This is an expected finding given

that selective, public institutions tend to be land-grant institutions or large public research

universities, which in turn are well resourced for research (Mumper, Gladieux, King, & Corrigan,

2011). Students at less selective institutions alternatively are more likely to join a club or

organization related to their major, which may be indicative of students placing a heavier

reliance on informal sources of support and professional development in the face of fewer

formal opportunities to support their success in STEM. Taken together, these findings support

the assertion that context matters with respect to student engagement in the opportunity

structure.

This study extends research on STEM persistence by identifying the educational

experiences and institutional contexts that contribute to student access and involvement in the

opportunity structure. Such research is especially timely given the inequitable educational

outcomes of certain groups of students – namely underrepresented racial minority students – in

STEM education. An overarching finding is that students who participate in one aspect of the


opportunity structure are also more likely to participate in other opportunities. The fact that the

advantages and benefits associated with the activities in the opportunity structure accrue

(Hurtado, Eagan, Cabrera, Lin, Park, & Lopez, 2008), underscores the need for early student

access to and participation in these experiences. Further, although it is ultimately up to

individual students to decide to become involved in those activities within the opportunity

structure, it is the institution’s responsibility to first create campuses that “are ripe with

opportunities for students to be engaged” (Wolf-Wendel et al., 2009, p. 425) and to actively and

aggressively encourage students from vulnerable backgrounds to participate. To this end,

institutions should early on disseminate information widely that informs students about such

opportunities, outlines the career and educational benefits resulting from involvement, and

ensures that students have the navigational skill to equitably take advantage of the opportunities

and become involved. In this way STEM educators can better ensure that all students who

aspire to pursue study in a STEM field, especially those who stand to benefit the most, receive

varied and multiple forms of exposure to activities that can help them succeed academically,

develop their talent, and secure a future in STEM.


References

Adelman, C. (2006). The Toolbox Revisited: Paths to Degree Completion From High School Through College. U.S. Department of Education. Washington, DC: Office of Vocational and Adult Education.

Allison, P. D. (2002). Missing data. Thousand Oaks, CA: SAGE. Arendale, D. (1998). Increasing efficiency and effectiveness of learning for freshman college

students through Supplemental Instruction. In J. L. Higbee & P. L. Dwinell (Eds.), Developmental education: Preparing successful college students (pp. 185-197). Columbia, SC: National Resource Center for The First-Year Experience & Students in Transition, University of South Carolina.

Armstrong, L., Power, C., Coady, C., & Dormer, L. (2011). Video-based supplemental instruction: Creating opportunities for at-risk students undertaking engineering mathematics. Journal of Peer Learning, 4(1), 3-15.

Aronson, J. & Inzlicht, M. (2004). The ups and downs of attributional ambiguity: Stereotype vulnerability and the academic self-knowledge of African American college students. Psychological Science, 15(12), 829-836.

Astin, A. W. (1993). What matters in college? Four critical years revisited. San Francisco, CA: Jossey-Bass.

Barlow, A. E. L., & Villarejo, M. (2004). Making a difference for minorities: Evaluation of an educational enrichment program. Journal of Research in Science Teaching, 41(9), 861-881.

Becher, T. (1989). Academic tribes and territories: Intellectual enquiry and the culture of disciplines. Buckingham: Open University Press

Blat, C. M. & Nunnally, K. (2004). Successfully applying the supplemental instruction model to engineering and pre engineering. Frontiers in Education, 3, 12-17.

Bohlscheid J., & Clark, S. (2012). Career preparedness survey outcomes of food science graduates – A follow-up assessment. Journal of Food Science Education, 11, 8-15.

Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241-258). New York: Greenwood.

Bordieu, P. & Passerson, J. (1977). Reproduction in education, society, and culture. London: Sage.

Braxton, J. M. (2000). Reworking the student departure puzzle. Nashville: Vanderbilt University Press.

Chang, M.J., Hurtado, S., Sharkness, J., and Newman, C.B. (in press). What matters in college for retaining aspiring scientists and engineers.

Chapman, D. W., & Pascarella, E. T. (1983). Predictors of academic and social integration of college students. Research in Higher Education, 19, 295-322.

CIRP Construct Technical Report: 2010 Appendix. (2010). Retrieved March 20, 2013 from http://www.heri.ucla.edu/PDFs/constructs/Appendix2010.pdf

Cole, J.S. & Gonyea, R.M. (2010). Accuracy of self-reported SAT and ACT test scores: Implications for research. Research in Higher Education, 51, 305-319.

Coletti, K.B., Covert, M., DiMilla, P.A., Gianino, L., & Reisberg, R. (2012). Integrating supplemental instruction into freshman chemistry programs to support women in engineering. Washington, D.C.: American Society for Engineering Education.

Cruce, T.M. (2009). A note on the calculation and interpretation of the delta-p statistic for categorical independent variables. Research in Higher Education, 50(6), 608-622. doi:10.1007/s11162-009-9131-1

de Leeuw, J., & Meijer, E. (2008). Introduction. In de Leeuw, J. & Meijer, E. (Eds.), Handbook of multilevel analysis. New York: Springer.


Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977). Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, 39(1), 1-38.

Do, B., Zhao, Y., Trytten, D., & Lowe, A. W. (November 2006). “Getting an internship… I’m still trying to find that:” Asian American student experiences obtaining engineering internships. Paper presented at annual meeting for APERA in Hong Kong.

Durham, S.A., & Marshall, W. E. (2012). Enhancing a student’s engineering experiences through participation on student organizations. AC 2012-3169

Eagan, K., Figueroa, T., Hurtado, S., & Gasiewski, J. (2012). Faculty Accessibility Cues: Opening the Doors to Classroom Communication. Paper presented at the Annual Forum of the Association for Institutional Research, New Orleans, LA.

Eagan, M.K., Herrera, F.A., Garibay, J.C., Hurtado, S., & Chang, M.J. (May 2011). Becoming STEM Protégés: Factors Predicting the Access and Development of Meaningful Faculty-Student Relationships. Paper presented at the annual meeting for the Association for Institutional Research.

Eagan, M. K., Hurtado, S., Chang, M. J., Garcia, G. A., Herrera, F., & Garibay, J. C. (in press). Making a Difference in Science Education for Underrepresented Students: The Impact of Undergraduate Research Programs.

Elliott, R., Strenta, A.C., Adair, R., Matier, M., & Scott, J. (1996). The role of ethnicity in choosing and leaving science in highly selective institutions. Research in Higher Education, 37(6), 681-709.

Espinosa, L. L. (2011). Pipelines and Pathways: Women of Color in Undergraduate STEM Majors and the College Experiences That Contribute to Persistence. Harvard Educational Review, 81(2), 209-241.

Gladieux, L. E., & Swail, W. S. (1998). Financial Aid is Not Enough: Improving the Odds of College Success. No185. Washington D.C.: Educaitonal Policy Institute.

Green, S. G., & Bauer, T. N. (1995). Supervisory mentoring by advisers: Relationships with doctoral student potential, productivity, and commitment. Personnel Psychology, 48, 537-561.

Hamrick, F. A., Schuh, J. H., & Shelley, M. C. (2004, May 4). Predicting higher education graduation rates from institutional characteristics and resource allocation. Education Policy Analysis Archives, 12, 19. Retrieved March 20, 2013 from http://epaa.asu.edu/ojs/article/view/174

Hands, D., Reid, I. and Younger, P. (1997). Video-based Supplemental Instruction for Engineering subjects and students at risk. Advancing International Perspectives, Proceedings of the HERDSA Annual International Conference, Adelaide. Retrieved March 12, 2013 from http://www.herdsa.org.au/wp-content/uploads/conference/1997/hands01.pdf

Hathaway, R.S., Nagda, B.R.A., & Gregerman, S.R. (2002). The relationship of undergraduate research participation to graduate and professional education pursuit: an empirical study. Journal of College Student Development, 43(5), 614-631.

Henderson, S. & Broadbridge, P. (December 2007). Mathematics for 21st century engineering students. Australasian Association for Engineering Education (AaeE) conference, 9-13, Melbourne, Australia. Retrieved March 25, 2013 from http://ww2.cs.mu.oz.au/aaee2007/papers/inv_Hend.pdf

Herrera, F., Hurtado, S., & Chang, M. (November 2011). Maintaining career aspirations in science, technology, engineering, and mathematics (STEM) among college students. Paper presented at the annual conference of the Association for the Study of Higher Education, Charlotte, NC.

Higher Education Research Institute (2010). Degrees of success: Bachelor’s degree completion rates among initial STEM majors. Retrieved on March 1, 2010, from

http://epaa.asu.edu/ojs/article/view/174


http://www.heri.ucla.edu/nih/downloads/2010%20-%20Hurtado,%20Eagan,%20Chang%20-%20Degrees%20of%20Success.pdf

Hunter, A., Laursen, S. L., & Seymour, E. (2007). “Becoming a Scientist: The Role of Undergraduate Research in Students’ Cognitive, Personal, and Professional Development.” Science Education, 91(1), 36–74.

Hurtado, S., & Carter, D.F. (1997). Effects of college transition and perception of the campus racial and social climate on Latino students’ sense of belonging. Sociology of Education, 70(4), 324-345.

Hurtado, S., Eagan, M. K., Cabrera, N. L., Lin, M. H., Park, J., & Lopez, M. (2008). Training future scientists: Predicting first-year minority student participation in health science research. Research in Higher Education, 49(2), 126-152.

Hurtado, S., Eagan, K., & Hughes, B. (2012). Priming the Pump or the Sieve: Institutional Contexts and URM STEM Degree Attainments. Paper presented at the Annual Forum of the Association for Institutional Research, New Orleans, LA.

Hurtado, S., Han, J. C., Sáenz, V. B., Espinosa, L., Cabrera, N., & Cerna, O. (2007). Predicting transition and adjustment to college: Biomedical and behavioral science aspirants’ and minority students’ first year of college. Research in Higher Education, 48(7), 841-887.

Institute of Electrical and Electronics Engineers (IEEE). (2007, November 18). IEEE position paper on the first professional degree in engineering. IEEE. Retrieved from www.ieee.org

Inroads. (1993). Partnership for a diversified workforce. St. Louis. MO: Inroads. Inzlicht, M. & Good, C. (2006). How Environments Can Threaten Academic Performance, Self-

Knowledge, and Sense of Belonging. In S. Levin & C. van Laar (Eds.), Stigma and group inequality: Social psychological perspectives (pp. 129-150). Mahwah, NJ: Lawrence Erlbaum Associates Publishers.

Jaeger, A. J., Eagan, M. K., & Wirt, L. G. (2008). Retaining students in science, math, and engineering majors: Rediscovering cooperative education. Journal of Cooperative Education and Internships, 42(1), 20-32.

Johnson, D. R. (2012). Campus Racial Climate Perceptions and Overall Sense of Belonging Among Racially Diverse Women in STEM Majors. Journal of College Student Development, 53(2), 336-346.

Jones, M. T., Barlow, A. E., & Villarejo, M. (2010). Importance of undergraduate research for minority persistence and achievement in biology. The Journal of Higher Education, 81(1), 82-115.

Kardash, C.M. (2000). Evaluation of an undergraduate research experience: Perceptions of undergraduate interns and their faculty mentors. Journal of Educational Psychology, 92(1), 191-201.

Kinkead, J. (2003). Learning through inquiry: An overview of undergraduate research. In J. Kinkead, (Ed.), Valuing and supporting undergraduate research (pp. 5-17). New Directions for Teaching and Learning, No. 93.

Kram, K. E. (1983). Phases of the mentor relationship. Academy of Management journal, 26(4), 608-625.

Lewis, J. B. (2001). Social justice, social studies, and social foundations. The Social Studies, 92(5), 189–192.

Link, T. (2003). Why students stay: Retention and persistence factors. Pennsylvania ABLE Administrators Field Notes. Retrieved March 1, 2013 from http://www.google.com.pr/url?sa=t&rct=j&q=why%20students%20stay%3A%20retention%20and%20persistence%20factors.%20&source=web&cd=1&ved=0CDYQFjAA&url=http%3A%2F%2Fwww.portal.state.pa.us%2Fportal%2Fserver.pt%2Fdocument%2F884402

http://www.ieee.org/


%2Ffn03retention_pdf&ei=PT1eUdGBJYrM9ASJxoHgDw&usg=AFQjCNFo-j5sN0VHwooW9uwSx9fFVrFFvA

Lopatto, D. (2003). The essential features of undergraduate research. Council on Undergraduate Research Quarterly, 23(1), 139-142.

Lopatto, D. (2004). Survey of Undergraduate Research Experiences (SURE): First findings. Cell Biology Education, 3(4), 270-277.

Malm, J., Bryngfors, L. E., & Mörner, L.L. (2010). Supplemental instruction (SI) at the Faculty of Engineering (LTH), Lund University, Sweden. An evaluation of the SI-program at five LTH engineering programs autumn 2008. Journal of Peer Learning, 3(1), 38-50.

Massey, D. S., Charles, C. Z., Lundy, G. F., and Fischer, M .J. (2002). The Source of the River: The Social Origins of Freshmen at America’s Selective Colleges and Universities. Princeton, NJ: Princeton University Press.

May, G.S., & Chubin, D.E. (2003). A retrospective on undergraduate engineering success for underrepresented minority students. The Research Journal for Engineering Education, 92(1), 27-39.

McLachlan, G. J., & Krishnan, T. (1997). The EM algorithm and extensions. New York: Wiley. Merkel, C. A. (2003). Undergraduate research at the research universities. New Directions for

Teaching and Learning, 93, 39-54. Mumper, M., Gladieux, L.E., King, J.E., & Corrigan, M.E. (2011). The federal government and

higher education. In P.G. Altbach, P.J. Gumport, & R.O. Berdahl (Eds.), American Higher Education in the Twenty-First Century (3rd Ed.). Baltimore, MD: Johns Hopkins University Press.

National Academy of Sciences (2011). Expanding underrepresented minority participation Washington, D.C.: The National Academies Press. Retrieved February 5, 2013 from http://grants.nih.gov/training/minority_participation.pdf

Outcalt, C. L. & Skewes-Cox, T. E. (2002). Involvement, interaction, and satisfaction: The human environment at HBCUs. The Review of Higher Education, 25(3), 331-347.

Packard, B. W. L. (2004). Mentoring and retention in college science: Reflections on the sophomore year. Journal of College Student Retention: Research, Theory and Practice, 6(3), 289-300.

Pascarella, E. T., & Terenzini, P. T. (2005). How college affects students: Vol. 2. A third decade of research. San Francisco, CA: Jossey-Bass.

Pender, M., Marcotte, D.E., Sto. Domingo, M.R., & Maton, K.I. (2010). The STEM pipeline: The role of summer research experience in minority students’ Ph.D. aspirations. Education Policy Analysis Archive, 18(30), 1-36.

Perna, L. W., Cooper, M., & Li, C. (2007). “Improving Educational Opportunities for Students Who Work. In E. P. St. John (ed.), Confronting Educational Inequality: Reframing, Building Understanding, and Making Change. New York: AMS Press.

Peterfreund, A. R., Rath, K. A., Xeos, S. P., & Bayliss, F. (2008). The impact of supplemental instruction on students in STEM courses: Results from San Francisco State University. Journal of College Student Retention, 9(4), 487-503.

Petersen, T. (1985). A comment on presenting results from logit and probit models. American Sociological Review, 50, 130-131.

Porter, S. R. (2006). Institutional structures and student engagement. Research in Higher Education, 47(5), 521-558.

President’s Council of Advisors on Science and Technology (PCAST). (2010). Prepare and inspire: K-12 education in science, technology, engineering, and math (STEM) for America’s Future. Washington, D.C.: Executive Office of the President. Retrieved February 1, 2013 from www.whitehouse.gov/sites/default/files/microsites/ostp/pcast-stemed-report.pdf.

http://www.whitehouse.gov/sites/default/files/microsites/ostp/pcast-stemed-report.pdf

http://www.whitehouse.gov/sites/default/files/microsites/ostp/pcast-stemed-report.pdf


Ragins, B. R. (1999). Gender and mentoring relationships: A review and research agenda for the next decade. In G. Powell (Ed.), Handbook of gender and work (pp. 347-370). Thousand Oaks, CA: Sage.

Ragins, B. R., & Cotton, J. L. (1993). Gender and willingness to mentor in organizations. Journal of Management, 19, 97-111.

Rath, K. A., Peterfreund, A. R., Xenos, S. P, Bayliss, F., & Carnal, N. (2007). Supplemental instruction in introductory biology I: Enhancing the performance and retention of underrepresented minority students. CBE-Life Sciences Education, 6, 203-216.

Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage Publishing.

Riggert, S. C., Boyle, M., Petrosko, J. M., Ash, D., & Rude-Parkins, C. (2006). Student employment and higher education: Empiricism and contradiction. Review of Educational Research, 76, 63–92.

Russell, S. H., Hancock, M. P., & McCullough, J. (2007). Benefits of undergraduate research experiences. Science, 316, 548 –549.

Rylands, L.J. & Coady, C. (2009). Performance of weak students in first-year mathematics and science. International Journal of Mathematical Education in Science and Technology, 40(6), 741-753.

Sagen H.B., Dallam J.W., & Laverty J.R. 2000. Effects of career preparation experiences on the initial employment success of college graduates. Research in Higher Education, 41(6), 753–67.

Schneider, B. (2000). Explaining the unrealized aspirations of racial and ethnic minorities. In G. Campbell, Jr., R. Denes & C. Morrison (Eds.), Access denied: Race, ethnicity, and the scientific enterprise (pp. 174-187). Oxford: Oxford University Press.

Schwitzer, A. M., Griffin, O. T., Ancis, J. R., & Thomas, C. R. (1999). Social adjustment experiences of African American college students. Journal of Counseling and Development, 77, 189-197.

Seymour, E., Hunter, A., Laursen, S.L., & Deantoni, T. (2004). Establishing the benefits of research experiences for undergraduates in the sciences: First findings from a three-year study. Science Education, 88(4), 493-534.

Seymour, E., & Hewitt, N. (1997). Talking about leaving: Why undergraduates leave the sciences. Boulder, CO: Westview Press.

Sharkness, J., DeAngelo, L., & Pryor, J. (2010). CIRP Construct Technical Report. Los Angeles, CA: Higher Education Research Institute. Retrieved March 20, 2013 from http://www.heri.ucla.edu/PDFs/constructs/technicalreport.pdf

Singh, R., Ragins, B. R., & Tharenou, P. (2009). Who gets a mentor? A longitudinal assessment of the rising star hypothesis. Journal of Vocational Behavior, 74(1), 11-17.

Stanton-Salazar, R. D. (1997). A social capital framework for understanding the socialization of racial minority children and youths. Harvard Education Review, 67(1), 1-40.

Stanton-Salazar, R. D. (2001). Manufacturing hope and despair: The school and kin support networks of U.S.-Mexican youth. New York: Teachers College Press.

Stanton-Salazar, R. D. (2010). A social capital framework for the study of institutional agents and their role in the empowerment of low-status students and youth. Youth & Society. Advance online publication. doi: 10.1177/0044118X10382877

Strayhorn, T.L. (2010). Undergraduate research participation and STEM graduate degree aspirations among students of color. In S.R. Harper & C.B. Newman (Eds.), Students of color in STEM (pp. 85-93). New Directions for Institutional Research, No. 148.

Thiry, H, Laursen, S. L., & Hunter, A. (2011). What experiences help students become scientists?: A comparative study of research and other sources of personal and


professional gains for STEM undergraduates. The Journal of Higher Education, 82(4), 357-388.

Tierney, W. G. (1992). An anthropological analysis of student participation in college. Journal of Higher Education 63(6), 603-607.

Tinto, V. (1993). Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). Chicago: University of Chicago Press.

Torres, V., Gross, J.P, & Dadashova, A. (2010).Traditional-aged students becoming at-risk: Does working threaten college students’ academic success? Journal of College Student Retention, 12(1), 51-68.

Tsang, E., Halderson, C., Abdel-Qader, I., Aller, B., Butt, S., Kline, A., ... & Kallen, K. (2006, October). Assessment of Faculty Mentoring Strategies of Student Learning Communities at Western Michigan University College of Engineering and Applied Sciences. In Frontiers in Education Conference, 36th Annual (pp. 1-6). IEEE.

Walton, G.M., & Cohen. G.L. (2007). A question of belonging: Race, social fit, and achievement. Journal of Personality and Social Psychology, 92(1), 82-96.

Wanberg, C. R., Welsh, E. T., & Hezlett, S. A. (2003). Mentoring research: A review and dynamic process model. Research in Personnel and Human Resources Management, 21, 39-124.

Wilson, S., Zakiya, S., Holmes, L., deGravelles, K., Sylvain, M.R., Batiste, L., Johnson, M., McGuire, S.Y., Pang, S.S. & Warner, I. (2011). Hierarchical Mentoring: A Transformative Strategy for Improving Diversity and Retention in Undergraduate STEM Disciplines, Journal of Science Education and Technology. DOI: 10.1007/s10956-011-9292-5

Wolf-Wendel, L., Ward, K., Kinzie, J. (2009). A tangled web of terms: The overlap and unique contribution of involvement, engagement, and integration to understanding college student success. Journal of College Student Development, 50(4), 407-428


Table 1. Descriptive statistics n=4046 students; n= 212 institutions

Variable Mean S.D. Min. Max

Dependent Variables

Participated in an internship program 0.46 0.5 0 1

Participated in an undergraduate research

program (e.g. MARC, MBRS, REU) 0.2 0.4 0 1

Joined a club or organization related to

major 0.6 0.49 0 1

Had instruction that supplemented course

work 2.14 0.64 1 3

Faculty mentorship 48.43 9.62 27.33 66.99

Demographic Characteristics

English native language 0.85 0.36 0 1

Gender (Female) 0.62 0.48 0 1

Native American 0.01 0.09 0 1

Asian American 0.12 0.32 0 1

Black 0.15 0.36 0 1

Latino 0.16 0.36 0 1

White 0.4 0.49 0 1

Other 0.01 0.09 0 1

Multiracial 0.15 0.36 0 1

Socioeconomic Status (SES) 0 1.1 -2.85 1.72

Pre-college preparation, achievement and experiences (Responses taken from 2004 TFS)

High School GPA 6.96 1.19 1 8

Years of Mathematics in H.S. 6.01 0.53 1 7

Years of Physical Science in H.S. 4.03 1.26 1 7

Years of Biological Science in H.S. 3.79 1.04 1 7

Participated in a summer research program 0.13 0.33 0 1

Participated in a health science research

program sponsored by a university 0.06 0.24 0 1

Enrolled in college to prepare for

graduate/professional school 2.69 0.58 1 3.12

Hours per week: Talking with high school

teachers outside class 2.68 1.01 1 8

Concerns about ability to finance college

education 1.87 0.64 1 3

Parent Occupation In STEM 0.33 0.47 0 1

SAT composite score 1201.68 181.46 500 1600

Degree aspiration: Less than bachelors

degree 0.01 0.07 0 1


Degree aspiration: Bachelors degree 0.09 0.29 0 1

Degree aspiration: Masters 0.27 0.44 0 1

Degree aspiration: Ph.D. or Ed.D. 0.32 0.47 0 1

Degree aspiration: MD 0.3 0.46 0 1

Degree aspiration: Other professional degree

(law, divinity, other) 0.02 0.14 0 1

College Experiences (Responses taken from the 2008 CSS)

Worked on independent study projects 2.12 0.77 1 3

Discussed course content with students

outside of class 2.67 0.51 1 3

Studied with other students 2.42 0.58 1 3

Tutored another college student 1.72 0.69 1 3

Met with an advisor/counselor about your

career plans 2.01 0.6 1 3

Asked a professor for advice outside of class 2 0.64 1 3

Felt intimidated by your professors 1.66 0.61 1 3

Felt isolated from campus life 1.62 0.66 1 3

Failed one or more courses 0.31 0.46 0 1

Worked full-time while attending school 0.19 0.39 0 1

Enrolled in honors or advanced courses 0.38 0.49 0 1

Participated in a program to prepare for

graduate school 0.19 0.39 0 1

Participated in an academic program for

racial/ethnic minorities 0.17 0.37 0 1

Presented research at a conference 0.18 0.39 0 1

Hours per Week: Socializing with friends 5.14 1.48 1 8

Hours per Week: Talking with faculty

outside of class or office hours 2.1 1.03 1 8

Hours per Week: Working for pay off

campus 3.17 2.75 1 8

Hours per Week: Commuting 2.58 1.48 1 8

Hours per Week: Career planning (job

searches, internships, etc.) 2.71 1.33 1 8

Faculty here are interested in students'

personal problems 2.63 0.75 1 4

There is strong competition among most

students for high grades 2.86 0.79 1 4

Faculty here are interested in students'

academic problems 2.99 0.68 1 4

Satisfaction: Courses in your major field 4.16 0.86 1 5


Satisfaction: Racial/ethnic diversity of the

student body 3.56 1.02 1 5

Career Concern: Discovery/enhancement of

knowledge 3.19 0.8 1 4

Overall College GPA 5.66 1.66 1 8

Academic Disengagement 52.36 8.08 36.29 75.71

Academic Self-Concept 51.58 8.58 12.65 66.92

Negative Cross-Racial Interactions 53.13 7.73 41.66 76.57

Positive Cross-Racial Interactions 53.89 8.6 29.06 68.39

Sense of Belonging 49.86 8.47 25.97 62.22

Social Self-Concept 49.87 8.55 19.36 67.26

Biological Sciences Major 0.28 0.45 0 1

Engineering Major 0.17 0.38 0 1

Professional Health Major 0.08 0.27 0 1

Math Major 0.02 0.13 0 1

Physical Science Major 0.05 0.23 0 1

Other Stem Major 0.08 0.28 0 1

Switched out of STEM into a non-STEM

major 0.4 0.49 0 1

Institutional Variables

Historically Black College/University (vs.

non-HBCU) 1.09 0.29 1 2

Institution offers a medical degree (vs. not) 1.23 0.42 1 2

Percentage of STEM undergraduates 0.16 0.15 0 0.89

Percent White undergraduates 0.58 0.25 0 0.96

Undergraduate full-time enrollment 8087.02 7259.43 404.5 36730.5

Institutional control: Public (vs. private) 1.57 0.5 1 2

Masters granting institution 0.36 0.48 0 1

Liberal arts institution 0.19 0.39 0 1

Doctoral granting institution 0.45 0.5 0 1

Selectivity (100-point increments) 11.14 1.46 7.75 15.1

Student peer mean: Faculty here are

interested in students' personal problems 2.71 0.36 1.6 4


Table 2. Results of hierarchical models predicting frequency of engagement in supplemental instruction

Model 1 Model 2 Model 3 Model 4

Variables b SE Sig. b SE Sig. b SE Sig. b SE Sig.

Level 1 (n=4046)

Model 1

English native language -.11 .03 ** -.10 .04 ** -.10 .04 ** -.09 .03 **

Gender (Female) .04 .02 .03 .03 .02 .02 .02 .02

Native American (reference is White) -.06 .16 -.06 .15 -.14 .11 -.15 .12

Asian American -.04 .04 -.05 .04 -.01 .03 -.01 .04

Black .02 .04 .04 .04 .01 .04 .03 .04

Latino -.05 .04 -.06 .04 -.06 .04 -.06 .04

Other -.09 .14 -.09 .14 -.17 .12 -.17 .12

Multiracial .01 .03 .00 .03 -.01 .03 -.01 .03

Socioeconomic Status (SES) .05 .01 ** .03 .01 * .02 .01 .02 .01

Model 2

High School GPA .02 .01 .00 .01 .00 .01

Years of Mathematics in H.S. .01 .02 .00 .02 .00 .02

Years of Physical Science in H.S. .00 .01 -.01 .01 -.01 .01

Years of Biological Science in H.S. -.01 .01 -.01 .01 -.01 .01

Participated in a summer research program .07 .03 * .06 .03 * .06 .03 *

Participated in a health science research program

sponsored by a university -.04 .05 -.07 .04 -.07 .04


graduate/professional school .03 .02 -.01 .02 -.01 .02

Hours per week: Talking with high school teachers

outside class .04 .01 ** .00 .01 .00 .01

Concerns about ability to finance college education -.01 .02 .00 .02 .00 .02


Parent Occupation In STEM .03 .03 .03 .03 .03 .03

SAT composite score (100) .02 .01 .01 .01 .00 .01

2004 Degree aspiration: Less than bachellors degree

(reference is bachellors degree) -.23 .19 -.06 .18 -.07 .18

2004 Degree aspiration: Masters -.05 .04 -.04 .04 -.04 .04

2004 Degree aspiration: Ph.D. or Ed.D. .02 .05 .01 .05 .00 .05

2004 Degree aspiration: MD .01 .05 -.01 .05 -.01 .05

2004 Degree aspiration: Other professional degree (law,

divinity, other) -.01 .08 -.01 .08 -.02 .08

Model 3

Worked on independent study projects .08 .02 ** .08 .02 **

Discussed course content with students outside of class .02 .02 .02 .02

Studied with other students .08 .02 ** .08 .02 **

Tutored another college student .05 .02 ** .05 .02 **

Met with an advisor/counselor about your career plans .03 .02 .02 .02

Asked a professor for advice outside of class .07 .02 ** .07 .02 **

Felt intimidated by your professors .03 .02 .03 .02

Felt isolated from campus life .02 .02 .02 .02

Failed one or more courses .01 .03 .00 .03

Worked full-time while attending school .00 .03 .00 .03

Enrolled in honors or advanced courses .02 .02 .02 .02

Participated in an internship program .00 .02 .00 .02

Participated in an undergraduate research program (e.g. MARC,

MBRS, REU) .02 .03 .02 .03

Participated in a program to prepare for graduate school -.02 .03 -.03 .03

Participated in an academic program for racial/ethnic minorities .07 .03 * .06 .03

Joined a club or organization related to major .06 .02 * .06 .02 **

Presented research at a conference -.04 .03 -.03 .03

Hours per Week: Socializing with friends .01 .01 * .02 .01 *

Hours per Week: Talking with faculty outside of class or office hours .02 .01 .02 .01


Hours per Week: Working for pay off campus .00 .00 .00 .00

Hours per Week: Commuting .00 .01 -.01 .01

Hours per Week: Career planning (job searches, internships, etc.) .00 .01 .00 .01

Faculty here are interested in students' personal problems .00 .02 -.01 .02

There is strong competition among most students for high grades .01 .01 .01 .01

Faculty here are interested in students' academic problems .04 .02 * .05 .02 **

Satisfaction: Courses in your major field .03 .02 * .03 .02

Satisfaction: Racial/ethnic diversity of the student body .01 .01 .01 .01

Career Concern: Discovery/enhancement of knowledge .00 .01 .00 .01

Overall College GPA .01 .01 .00 .01

Academic Disengagement (factor) (10) .03 .01 * .03 .01 *

Academic Self-Concept (factor) (10) .02 .01 .02 .01

Negative Cross-Racial Interactions (factor) (10) -.01 .02 -.01 .02

Positive Cross-Racial Interactions (factor) (10) .05 .02 ** .05 .02 **

Sense of Belonging (factor) (10) .05 .02 ** .05 .02 **

Social Self-Concept (factor) (10) -.01 .01 -.01 .01

Faculty Mentorship (factor) (10) .07 .02 ** .07 .02 **

Biological Sciences Major (reference is students who

switched to non-STEM majors) -.02 .03 -.02 .03

Engineering Major -.04 .04 -.04 .04

Professional Health Major .00 .05 .01 .05

Math Major -.17 .10 -.16 .10

Physcial Science Major .05 .05 .05 .05

Other Stem Major .04 .04 .04 .04

Model 4

Level 2 (n=212)

Intercept 2.22 .04 ** 2.22 .06 ** 2.19 .06 ** 2.06 .11 **

HBCU (vs. non-HBCU) .01 .07

Institution offers a medical degree (vs not) .02 .03


Proportion of STEM undergraduate majors .02 .10

Proportion of undergraduate White students .04 .07

Undergraduate full-time enrollment (1000) .00 .00

Private (vs. public) .01 .04

Masters comprehensive institution (vs. liberal arts) .03 .05

Research university (vs. liberal arts) .06 .05

Selectivity (100-point increments) -.01 .02

Student peer mean: Faculty here are interested in students' personal problems -.03 .05

% Level-1 variance explained 0.68% 1.80% 16.11% 5.74%

% Level-2 variance explained n/a n/a n/a 51.93%

*indicates p-value less than .05; **indicates p-value less than .01


Table 3. Results of hierarchical models predicting faculty mentorship score

Model 1 Model 2 Model 3 Model 4

Variables β SE Sig. β SE Sig. β SE Sig. β SE Sig.

Level 1 (n=4046)

Model 1

English native language -.58 .57 -.40 .58 .26 .40 .09 .40

Gender (Female) 1.28 .39 ** .76 .41 .52 .29 .52 .28

Native American (reference is

White) 1.42 2.32 .75 2.37 1.20 1.58 1.17 1.58

Asian American -1.27 .62 * -1.35 .61 * -.37 .45 -.25 .44

Black -.61 .57 -.76 .60 -.27 .43 -.76 .48

Latino -.58 .64 -1.15 .64 -.47 .39 -.49 .39

Other .15 2.17 .14 2.33 -1.28 1.93 -1.13 1.89

Multiracial -1.51 .51 ** -1.59 .49 ** -.97 .34 ** -1.04 .33 **

Socioeconomic Status (SES) .47 .15 ** .39 .16 * .23 .13 .22 .13

Model 2

High School GPA .47 .17 * -.11 .12 -.07 .12

Years of Mathematics in H.S. .13 .33 .11 .23 .14 .23

Years of Physical Science in H.S. .06 .13 -.09 .09 -.07 .09

Years of Biological Science in H.S. .03 .16 .03 .11 .03 .11

Participated in a summer research program -.24 .49 -.34 .38 -.34 .38


sponsored by a university .96 .59 -.30 .53 -.43 .54


graduate/professional school 1.21 .37 ** .29 .23 .34 .23


teachers outside class 1.24 .16 ** .29 .12 * .28 .12 *

Concerns about ability to finance college education -.79 .27 ** -.49 .20 ** -.46 .20 *

Parent Occupation In STEM -.46 .37 -.37 .25 -.29 .25


SAT composite score (100) -.21 .13 -.36 .11 ** -.27 .12 *

2004 Degree aspiration: Less than bachelors

degree (reference is a bachelors degree) -6.59 1.16 ** -2.52 1.27 * -2.57 1.16 *

2004 Degree aspiration: Masters .28 .60 .80 .42 .83 .42 *

2004 Degree aspiration: Ph.D. or Ed.D. .11 .67 .03 .45 .03 .45

2004 Degree aspiration: MD .30 .77 .18 .56 .12 .56

2004 Degree aspiration: Other professional degree (law,

divinity, other) .99 1.38 .53 1.39 .45 1.32

Model 3

Worked on independent study projects .86 .16 ** .85 .15 **

Discussed course content with students outside of class .43 .28 .44 .27

Studied with other students .36 .24 .32 .23

Tutored another college student .14 .20 .11 .20

Met with an advisor/counselor about your career plans 1.55 .21 ** 1.59 .21 **

Asked a professor for advice outside of class 2.18 .22 ** 2.08 .22 **

Felt intimidated by your professors -.58 .22 ** -.58 .21 **

Felt isolated from campus life -.55 .21 ** -.56 .21 **

Had instruction that supplemented course work .88 .21 ** .90 .21 **

Failed one or more courses -.49 .34 -.48 .34

Worked full-time while attending school -.40 .32 -.48 .31

Enrolled in honors or advanced courses .60 .30 * .57 .29 *

Participated in an internship program .33 .25 .39 .24

Participated in an undergraduate research program (e.g. MARC, MBRS,

REU) .52 .36 .67 .35

Participated in a program to prepare for graduate school 1.39 .32 ** 1.30 .32 **

Participated in an academic program for racial/ethnic minorities .06 .32 .13 .32

Joined a club or organization related to major .40 .26 .30 .25

Presented research at a conference 1.28 .39 ** 1.16 .38 **

Hours per Week: Socializing with friends -.07 .10 -.09 .10


Hours per Week: Talking with faculty outside of class or office hours 1.43 .13 ** 1.40 .13 **

Hours per Week: Working for pay off campus .05 .05 .04 .05

Hours per Week: Commuting .02 .09 .01 .09

Hours per Week: Career planning (job searches, internships, etc.) -.04 .11 -.03 .11

Faculty here are interested in students' personal problems 1.83 .18 ** 1.99 .18 **

There is strong competition among most students for high grades .02 .17 .15 .16

Faculty here are interested in students' academic problems 2.31 .21 ** 2.09 .21 **

Satisfaction: Courses in your major field 1.12 .17 ** 1.14 .16 **

Satisfaction: Racial/ethnic diversity of the student body .58 .14 ** .57 .14 **

Career Concern: Discovery/enhancement of knowledge .92 .16 ** .91 .15 **

Overall College GPA .35 .11 ** .31 .11 **

Academic Disengagement (factor) (10) -.47 .15 ** -.42 .15 **

Academic Self-Concept (factor) (10) .11 .18 .12 .18

Negative Cross-Racial Interactions (factor) (10) .29 .18 .28 .18

Positive Cross-Racial Interactions (factor) (10) .13 .16 .22 .16

Sense of Belonging (factor) (10) .34 .19 .28 .19

Social Self-Concept (factor) (10) -.07 .16 -.04 .15

Biological Sciences Major (reference is students who

switched to non-STEM majors) -.46 .32 -.62 .32

Engineering Major -1.54 .38 ** -1.35 .38 **

Professional Health Major .53 .52 .47 .53

Math Major -2.11 1.01 * -2.09 .99 *

Physical Science Major .44 .54 .39 .54

Other Stem Major -.67 .50 -.72 .49

Model 4

Level 2 (n=212)

Intercept 49.65 .70 ** 49.99 .91 ** 47.83 .63 ** 44.40 1.58 **

HBCU (vs. non-HBCU) 2.54 .78 **


Institution offers a medical degree (vs. not) .27 .41

Proportion of STEM undergraduate majors -1.89 .86 *

Proportion of undergraduate White students 2.23 .69 **

Undergraduate full-time enrollment (1000) -.06 .02 **

Private (vs. public) .13 .43

Masters comprehensive institution (vs. liberal arts) 1.03 .49 *

Research university (vs. liberal arts) .95 .51

Selectivity (100-point increments) -.31 .19

Student peer mean: Faculty here are interested in students' personal problems 2.87 .61 **

% Level-1 variance explained 0.57% 3.72% 50.00% 43.90%

% Level-2 variance explained n/a n/a n/a 93.67%



Table 4. Results of hierarchical models predicting student participation in an internship program, an undergraduate research program, and a club or

organization related to major

Participated in an internship

program

Participated in an

undergraduate research

program

Joined a club or

organization related to

major

Variables b SE Sig. Delta-P b SE Sig.

Delta-

P b SE Sig.

Delta-

P

Level 1 (n=4046)

Model 1

English native language -.01 .14 .95 .10 .16 .52 -.18 .13 .15

Gender (Female) .19 .10 .05 -.24 .12 .05 .17 .09 .08

Native American (reference: White) -.35 .40 .39 -1.91 1.20 .11 -.06 .46 .90

Asian American .05 .14 .70 .31 .21 .13 .01 .14 .95

Black .06 .14 .65 .89 .23 .00 16.27% -.27 .14 .06

Latino .19 .14 .15 .17 .20 .39 -.05 .15 .74

Other -.35 .48 .47 .15 .64 .82 .57 .46 .21

Multiracial .08 .12 .50 .13 .16 .43 .00 .16 .99

Socioeconomic Status (SES) .13 .05 .00 3.29% .00 .07 .94 .03 .05 .55

Model 2

High School GPA -.02 .04 .66 -.02 .06 .76 .17 .04 .00 3.98%

Years of Mathematics in H.S. -.03 .08 .72 .05 .10 .66 .16 .09 .06

Years of Physical Science in H.S. .00 .03 .91 -.10 .04 .01 -1.54% -.01 .03 .86

Years of Biological Science in H.S. .05 .04 .23 .01 .05 .88 -.01 .05 .78

Participated in a summer research program? .08 .12 .53 .13 .17 .46 -.13 .16 .41


sponsored by a university .03 .18 .88 .05 .24 .85 .66 .19 .00 14.47%


graduate/professional school -.21 .08 .01 -5.21% -.05 .12 .69 -.08 .09 .35


teachers outside class -.01 .04 .85 -.04 .05 .49 .06 .05 .23


Concerns about ability to finance college

education -.08 .07 .23 .07 .09 .45 -.06 .08 .41

Parent Occupation In STEM .06 .11 .58 -.29 .12 .02 -4.60% .00 .10 .98

SAT composite score (100) -.10 .04 .01 -2.53% .16 .06 .01 2.58% -.01 .04 .76

2004 Degree aspiration: Less than bachelors

degree (reference is bachelors degree) -.70 .69 .30 1.35 .86 .12 -1.21 .69 .08

2004 Degree aspiration: Masters .05 .16 .77 .46 .25 .07 .04 .15 .82

2004 Degree aspiration: Ph.D. or Ed.D. .16 .16 .30 .64 .25 .01 9.51% .11 .17 .51

2004 Degree aspiration: MD -.12 .17 .50 .45 .27 .09 .18 .20 .34

2004 Degree aspiration: Other professional

degree (law, divinity, other) .18 .36 .63 .10 .54 .85 .01 .40 .99

Model 3

Worked on independent study projects .01 .05 .87 .34 .08 .00 6.12% -.09 .05 .09

Discussed course content with students outside

of class .01 .10 .91 -.05 .13 .70 .30 .09 .00 6.87%

Studied with other students .12 .08 .14 -.13 .10 .22 .09 .08 .28

Tutored another college student .09 .08 .25 .09 .08 .25 .20 .07 .01 4.69%

Met with an advisor/counselor about your career

plans .14 .07 .06 .20 .11 .08 -.11 .07 .13

Asked a professor for advice outside of class .13 .08 .10 .04 .10 .67 .23 .07 .00 5.32%

Felt intimidated by your professors .06 .07 .41 .17 .11 .12 .00 .08 .97

Felt isolated from campus life .06 .07 .38 .12 .11 .26 .05 .08 .52

Had instruction that supplemented course work .01 .07 .90 .06 .10 .54 .17 .07 .02 4.07%

Failed one or more courses -.42 .10 .00 -10.42% -.02 .19 .90 .11 .11 .32

Worked full-time while attending school -.27 .12 .03 -6.56% -.55 .18 .00 -7.98% -.26 .12 .03 -6.36%

Enrolled in honors or advanced courses .29 .09 .00 7.18% .12 .13 .37 .24 .11 .03 5.66%

Participated in an internship program - - - .18 .13 .14 .45 .09 .00 10.77%

Participated in an undergraduate research

program (e.g. MARC, MBRS, REU) .23 .13 .07 - - - .15 .14 .28

Participated in a program to prepare for graduate

school .19 .12 .14 1.07 .13 .00 20.61% .33 .13 .01 7.77%


Participated in an academic program for

racial/ethnic minorities .40 .14 .00 10.05% .23 .13 .08 .37 .14 .01 8.63%

Joined a club or organization related to major .44 .09 .00 10.90% .23 .14 .09 - - -

Presented research at a conference .41 .12 .00 10.11% 1.71 .13 .00 35.02% .36 .12 .00 8.33%

Hours per Week: Socializing with friends .03 .03 .27 -.02 .04 .67 .00 .03 .94

Hours per Week: Talking with faculty outside of

class or office hours -.05 .04 .22 .04 .05 .48 .16 .06 .01 3.69%

Hours per Week: Working for pay off campus .04 .02 .01 1.03% -.01 .03 .64 .00 .02 .81

Hours per Week: Commuting .00 .03 .89 -.05 .04 .24 -.10 .03 .00 -2.37%

Hours per Week: Career planning (job searches,

internships, etc.) .29 .03 .00 7.25% -.05 .04 .25 .09 .03 .01 2.13%

Faculty here are interested in students' personal

problems .07 .07 .35 -.05 .10 .64 -.12 .07 .08

There is strong competition among most students

for high grades -.10 .06 .08 -.17 .08 .04 -2.60% .20 .06 .00 4.64%

Faculty here are interested in students' academic

problems -.18 .07 .01 -4.46% .29 .09 .00 5.06% .10 .08 .21

Satisfaction: Courses in your major field -.01 .05 .80 -.14 .07 .06 .05 .06 .38

Satisfaction: Racial/ethnic diversity of the

student body -.12 .04 .01 -2.87% .06 .06 .32 -.05 .05 .26

Career Concern: Discovery/enhancement of

knowledge -.04 .05 .38 .18 .08 .03 3.08% -.07 .06 .24

Overall College GPA .11 .04 .01 2.81% .18 .05 .00 3.02% .04 .03 .22

Academic Disengagement (factor) (10) -.08 .06 .19 -.05 .08 .50 .01 .05 .80

Academic Self-Concept (factor) (10) -.11 .06 .08 -.01 .09 .94 .03 .06 .59

Negative Cross-Racial Interactions (factor) (10) .10 .06 .09 .01 .09 .90 -.04 .07 .58

Positive Cross-Racial Interactions (factor) (10) -.07 .06 .20 .05 .08 .52 -.04 .07 .51

Sense of Belonging (factor) (10) .08 .07 .26 -.07 .09 .43 .15 .07 .03 3.58%

Social Self-Concept (factor) (10) .18 .07 .01 4.48% .01 .08 .86 .01 .06 .92

Faculty Mentoring (factor) (10) .09 .06 .12 .15 .09 .08 .06 .06 .33


Biological Sciences Major (reference is students

who switched to non-STEM majors) -.21 .12 .09 .78 .17 .00 13.89% -.02 .12 .87

Engineering Major 1.12 .13 .00 26.57% .16 .19 .41 .85 .14 .00 19.02%

Professional Health Major -.52 .21 .01 -12.05% -1.01 .35 .01 -9.93% .60 .20 .00 14.03%

Math Major -.42 .31 .18 -.10 .42 .82 -.50 .30 .09

Physical Science Major -.21 .22 .34 1.21 .25 .00 23.54% .20 .19 .29

Other Stem Major .31 .14 .03 7.71% .21 .26 .43 .16 .16 .34

Model 4

Level 2 (n=212)

Intercept -1.68 .54 .00 -3.63 .72 .00 -1.23 .55 .03

HBCU (vs. non-HBCU) -.09 .38 .81 .39 .50 .44 .69 .35 .05

Institution offers a medical degree (vs. not) -.02 .12 .88 .21 .19 .26 .15 .13 .26

Proportion of STEM undergraduate majors .60 .48 .21 .89 .53 .09 .39 .37 .30

Proportion of undergraduate White students -.18 .34 .61 -.42 .55 .44 .51 .36 .16

Undergraduate full-time enrollment (1000) .02 .01 .09 .00 .01 .76 .01 .01 .53

Private (vs. public) .57 .18 .00 13.89% -.52 .25 .04 -8.65% .17 .18 .34

Masters comprehensive institution (vs. liberal

arts) .04 .24 .86 -.07 .27 .80 -.01 .19 .95

Research university (vs. liberal arts) -.24 .25 .33 -.09 .29 .76 .09 .20 .67

Selectivity (100-point increments) .18 .07 .02 4.51% .28 .10 .01 4.55% -.32 .07 .00 -7.98%

Student peer mean: Faculty here are interested in

students' personal problems -.28 .27 .29 -.31 .34 .37 .31 .29 .28



Appendix A

Variables and coding

Variable Coding Scheme

Dependent Variables

Participated in an internship

program

0=no; 1=yes

Participated in an undergraduate

research program (e.g. MARC,

MBRS, REU)

0=no; 1=yes

Joined a club or organization related

to major

0=no; 1=yes

Had instruction that supplemented

course work 1= not at all; 2=occasionally; 3= frequently

Faculty mentorship Continuous; Nine-item factor (see Appendix

B)

Demographic Characteristics

English native language 0=no; 1=yes

Gender (Female) 0=male, 1=female

Native American 0=no; 1=yes

Asian American 0=no; 1=yes

Black 0=no; 1=yes

Latino 0=no; 1=yes

Other 0=no; 1=yes

Multiracial 0=no; 1=yes

Socioeconomic Status (SES)

Continuous; Factor of mother's education,

father's education, and parental income (see

Appendix B)

Pre-college preparation, achievement and experiences (Responses taken from TFS)

High School GPA 1=D to 8=A or A+

Years of Mathematics in H.S. 1=none to 7=5+ years

Years of Physical Science in H.S. 1=none to 7=5+ years

Years of Biological Science in H.S. 1=none to 7=5+ years

Participated in a summer research

program 0=no; 1=yes

Participated in a health science

research program sponsored by a

university 0=no; 1=yes


graduate/professional school 0=no; 1=yes

Hours per week: Talking with high

school teachers outside class 1=none to 8=Over 20 hours


Concerns about ability to finance

college education 1=none; 2=some; 3=major

Parent Occupation In STEM 0=no; 1=yes

SAT composite score Continuous

2004 Degree aspiration: Less than

bachelors degree (reference is a

bachelors degree) 0=no; 1=yes

2004 Degree aspiration: Masters 0=no; 1=yes

2004 Degree aspiration: Ph.D. or

Ed.D. 0=no; 1=yes

2004 Degree aspiration: MD 0=no; 1=yes

2004 Degree aspiration: Other

professional degree (law, divinity,

other) 0=no; 1=yes

College Experiences (Responses taken from the CSS)

Worked on independent study

projects 1= not at all; 2=occasionally; 3= frequently

Discussed course content with

students outside of class 1= not at all; 2=occasionally; 3= frequently

Studied with other students 1= not at all; 2=occasionally; 3= frequently

Tutored another college student 1= not at all; 2=occasionally; 3= frequently

Met with an advisor/counselor about

your career plans 1= not at all; 2=occasionally; 3= frequently

Asked a professor for advice outside

of class 1= not at all; 2=occasionally; 3= frequently

Felt intimidated by your professors 1= not at all; 2=occasionally; 3= frequently

Felt isolated from campus life 1= not at all; 2=occasionally; 3= frequently

Failed one or more courses 0=no; 1=yes

Worked full-time while attending

school 0=no; 1=yes

Enrolled in honors or advanced

courses 0=no; 1=yes

Participated in a program to prepare

for graduate school 0=no; 1=yes

Participated in an academic program

for racial/ethnic minorities 0=no; 1=yes

Presented research at a conference 0=no; 1=yes

Hours per Week: Socializing with

friends 1=none to 8=Over 20 hours

Hours per Week: Talking with

faculty outside of class or office

hours 1=none to 8=Over 20 hours


Hours per Week: Working for pay

off campus 1=none to 8=Over 20 hours

Hours per Week: Commuting 1=none to 8=Over 20 hours

Hours per Week: Career planning

(job searches, internships, etc.) 1=none to 8=Over 20 hours

Faculty here are interested in

students' personal problems

1= strongly disagree; 2= disagree; 3=agree;

4= Strongly disagree

There is strong competition among

most students for high grades



Faculty here are interested in

students' academic problems



Satisfaction: Courses in your major

field

1=Very dissatisfied; 2=dissatisfied;

3=neutral; 4=satisfied; 5=Very satisfied

Satisfaction: Racial/ethnic diversity

of the student body

1=Very dissatisfied; 2=dissatisfied;

3=neutral; 4=satisfied; 5=Very satisfied

Career Concern:

Discovery/enhancement of

knowledge

1= Not Important; 2= Somewhat important;

3= Very important Essential

Overall College GPA 1=D to 8=A or A+

Academic Disengagement

Continuous; Four-item factor (see Appendix

B)

Academic Self-Concept

Continuous; Five-item factor (see Appendix

B)

Negative Cross-Racial Interactions

Continuous; Three-item factor (see Appendix

B)

Positive Cross-Racial Interactions Continuous; Six-item factor (see Appendix B)

Sense of Belonging

Continuous; Four-item factor (see Appendix

B)

Social Self-Concept

Continuous; Three-item factor (see Appendix

B)

Biological Sciences Major 0=no; 1=yes

Engineering Major 0=no; 1=yes

Professional Health Major 0=no; 1=yes

Math Major 0=no; 1=yes

Physical Science Major 0=no; 1=yes

Other Stem Major 0=no; 1=yes

Institutional Variables

Intercept Institutional control: Public (vs. private)

Historically Black

College/University (vs. non-HBCU) 0=non-HBCU, 1=HBCU


Institution offers a medical degree

(vs. not) 0=no; 1=yes

Percentage of STEM undergraduates Continuous

Percent White undergraduates Continuous

Undergraduate full-time enrollment

(proxy for institutional size) Continuous

Institutional control: Public (vs.

private)

0=Public 1=Private

Masters granting institution (vs.

liberal arts) 0=no; 1=yes

Doctoral granting institution (vs.

liberal arts) 0=no; 1=yes

Selectivity (100-point increments) Continuous

Student peer mean: Faculty here are

interested in students' personal

problems Continuous


Appendix B

Factor and Construct Items

Factor Item

Cronbach's

Alpha

Factor

Loading

Socioeconomic Status (TFS) 0.731

Father's education 0.826

Mother's education 0.765

Parental income 0.595

Constructs (see CIRP, 2011)

Faculty Mentorship (CSS) - Measures the extent to which

students and faculty interact in relationships that foster

mentorship, support, and guidance, with respect to both academic

and personal domains.

Help in achieving professional goals

Advice and guidance about educational program

Encouragement to pursue graduate/professional study

A letter of recommendation

Feedback about academic work outside of grades

An opportunity to work on a research project

Academic Disengagement (CSS) - Measures the extent to which

students engage in behaviors that are inconsistent with academic

success.

Come late to class

Fell asleep in class

Failed to complete homework on time

Missed class for other reasons

Academic Self-Concept (CSS)- A unified measure of students'

beliefs about their abilities and confidence in academic

environments.

Academic ability

Self-confidence (intellectual)

Mathematical Ability

Writing ability

Negative Cross-Racial Interaction - A unified measure of

students' level of negative interactions with diverse peers.

Had guarded, cautious interactions

Had tense, somewhat hostile interactions

Felt insulted or threatened because of race/ethnicity

Positive Cross-Racial Interaction - A unified measure of students'

level of positive interactions with diverse peers.


Dined or shared a meal

Had meaningful and honest discussions about race/ethnic relations

outside of class

Shared personal feelings and problems

Had intellectual discussions outside of class

Studied or prepared for class

Socialized or partied

Sense of Belonging (CSS) - Measures the extent to which students

feel a sense of academic and social integration on campus.

I feel I have a sense of belonging to this campus

I feel I am a member of this college

I see myself as part of the campus community

Social Self-Concept (CSS)- A unified measure of students' beliefs

about their abilities and confidence in social situations.

Self-Confidence (social)

Leadership ability

Understanding of others


Appendix C

List of Majors Defined as STEM

1. General Biology

2. Biochemistry/Biophysics

3. Botany

4. Environmental Science

5. Marine (Life) Science

6. Microbiology/Bacterial Biology

7. Zoology

8. Other Biological Science

9. Aeronautical/Astronautical Engineering

10. Civil Engineering

11. Chemical Engineering

12. Computer Engineering

13. Electrical Engineering

14. Industrial Engineering

15. Mechanical Engineering

16. Other Engineering

17. Astronomy

18. Atmospheric Science

19. Chemistry

20. Earth Science

21. Marine Science

22. Mathematics

23. Physics

24. Statistics

25. Other Physical Science

26. Health Technology

27. Medicine/Dentistry/Veterinary Medicine

28. Nursing

29. Pharmacy

30. Agriculture

31. Computer Science

Documents

STEM Opportunity Structure 1 - HERI...STEM Opportunity Structure 4 their goals (Link, 2003), provide academic enrichment, offer a sense of guidance and support (Barlow & Villarejo,